Showing posts with label Wikipedia. Show all posts
Showing posts with label Wikipedia. Show all posts

Wednesday, April 6, 2011

Tireless Wikipedian Frederick Rhine

One of the co-editors of this blog gets well-deserved kudos at Dana Blogs Chess for his contributions to WikiProject: Chess.

Frederick and I rewrote large sections of Wikipedia's Bobby Fischer article a couple of years ago.  This reminds me that I need to write a review of Frank Brady's Endgame after tax season, as the article's weakness mirrors one of the biggest flaws in Brady's book (otherwise strongly recommended!): the coverage of Fischer's miracle years 1970-71 is embarrassingly thin.

Tuesday, September 21, 2010

Fun with Morphy Numbers

Ever hear of Bacon numbers? They are derived from the concept of Six Degrees of Separation. If you are the actor Kevin Bacon himself, you have a Bacon number of 0. If you played in a movie with him, you have a Bacon number of 1. If you played in a movie with someone who played in a movie with Bacon, you have a Bacon number of 2, and so on.

Bacon numbers were inspired by Erdős numbers, which Caspar Goffman introduced in 1969, 25 years before Bacon numbers. The brilliant, eccentric and prolific mathematician Paul Erdős authored or co-authored about 1400 published mathematical papers, more than anyone else in history. If you co-authored a paper with him, you have an Erdős number of one; if you co-authored a paper with someone who co-authored a paper with Erdős, you have an Erdős number of two, etc. Danica McKellar, who played Winnie Cooper on the TV series The Wonder Years (1988-93), has an Erdős number of 4 and a Bacon number of 2. For more on Erdős, I recommend the fascinating book The Man Who Loved Only Numbers: The Story of Paul Erdős and the Search for Mathematical Truth by Paul Hoffman, who later wrote King's Gambit: A Son, a Father, and the World's Most Dangerous Game.

Tim Krabbé was apparently the first to adapt the concept to chess, introducing Morphy numbers. Taylor Kingston then took up the cudgels in an article on chesscafe.com. Someone who played Morphy (in any sort of game, including an offhand game or simultaneous exhibition) has a Morphy number of 1 (MN1). Someone who played someone who played Morphy has a Morphy number of 2 (MN2), etc. For example, Emanuel Lasker played many games against Henry Bird, who played 12 games with Morphy in 1858 and 1859. Bird thus was an MN1, and Lasker an MN2. Anyone who played Lasker is an MN3.

As you might imagine, everyone who played Morphy died long ago. The lowest Morphy number held by anyone alive today is 3. Until very recently, it was thought that there were only a handful of living MN3s. Kingston in his article pointed out two: GMs Andor Lilienthal and Arturo Pomar. Lilienthal played R.P. Michell at Hastings 1934-35; Michell played the Rev. John Owen at Hastings 1895 (Amateur Section); and Owen played Morphy a number of games in 1858. Pomar played a match against Jacques Mieses in 1946; Mieses played Louis Paulsen at Nuremberg 1888 and Henry Bird at Hastings 1895; and both Paulsen and Bird played a number of games against Morphy. Leonard Barden, in his comments to another article about Morphy numbers, by Tim Harding, added three more living MN3s, all of whom played Mieses in the Premier Reserves section at Hastings 1948-49 or 1949-50: Dennis Horne, Peter Swinnerton-Dyer, and Barden himself.

Sadly, the great Lilienthal died earlier this year at age 99. That left only four living MN3s from the Kingston-Barden list. But wait, there's more! On April 28, 1938, 8-year-old Melvin Chernev (Irving's son), played a game against Emanuel Lasker (MN2) when the great man visited the Chernevs' home. Melvin Chernev lives near San Francisco; I spoke with him on the phone in June of this year. He is the fifth known living player with a Morphy number of 3.

Chicago's own Erik Karklins, age 95, recalls losing to Lasker in a simul in Riga, Latvia circa 1928. Karklins is thus the sixth known living player with a Morphy number of 3.

T.H. George of Essex, England may have been the last surviving player with a Morphy number of 2. The British Chess Magazine wrote in his 1971 obituary, "Essex chess players suffered a severe loss with the death of T.H. George on April 19th at the age of 85. He was the doyen of Essex chessplayers . . . . He was rather proud of having played a man who had played Morphy. This happened in his young days when he beat Jas. Mortimer in a club match; Mortimer had played friendly games with Morphy in Paris in the early sixties of the last century." British Chess Magazine, July 1971, p. 249. The claim that Mortimer played Morphy is corroborated by Chess Monthly, which had an article about Mortimer (1833-1911) in 1892. "In 1853, he was appointed attaché of the United States Legation in Paris, where he had an opportunity of renewing his acquaintance with Paul Morphy. The two countrymen thus became intimate friends. Both being passionately fond of chess, many hundreds of games were played by the master and pupil . . . ." Chess Monthly, Sept. 1892, p. 66.

Leonard Barden, in an e-mail to me yesterday morning, made a fascinating point:
Mortimer played Znosko-Borovsky and (most notably) Tartakover at Ostend 1907 which blows up Morphy 3 as a finite one-hand number[.]

Z-B played at London 1948 (two Penrose brothers), Tartakover met Gligoric, Benko, Bisguier, Fuderer (Bled 1950) , Matanovic, Ivkov, J Penrose and others[.] Tom George likely played Peter Clarke who was a young member of Ilford CC in the early 1950s when George was still active.

I reckon now the traceable living Morphy 3s number around 15-20.

Other opponents of Tartakower's who are still living include Fridrik Olafsson, Lothar Schmid, and possibly Gudbjartur Gudmundsson. In addition, Mortimer played Marshall at Monte Carlo 1902, and Marshall played Louis Levy in the 1941 Marshall Chess Club Championship. As far as I can tell from the USCF and FIDE websites, Levy is alive and living in California.

One wonders whether anyone alive today played Frederick Karl Esling (1860-1955). Esling, who became the first Australian champion in 1886, won an offhand game against Adolf Anderssen in 1878. Unfortunately, he does not seem to have played much in his later years; the last game of his that I have seen was from 1922. Since Anderssen of course played a match against Morphy, anyone who played Esling would have a Morphy number of 3 (and, incredibly, an Anderssen number of 2, even though Anderssen predeceased Morphy). UPDATE: According to Chess World, Vol. 5, No. 7 (July 1, 1950), page 156, Esling's last serious chess game occurred in 1924, when he played for Victoria in a match against New South Wales.

I played (and managed to draw) a 15-minute online game against Barden earlier this year, giving me a Morphy number of 4. But I also played two simul games against Bisguier around 1978, making me an MN4 through the Bisguier-Tartakower-Mortimer route as well.

In sum, here's an alphabetical list of known living MN3s: Leonard Barden, Pal Benko, Arthur Bisguier, Melvin Chernev, Andrija Fuderer, Svetozar Gligorić, Dennis Horne, Borislav Ivkov, Erik Karklins, Louis Levy, Aleksandar Matanović, Fridrik Olafsson, Jonathan Penrose, Oliver Penrose, Arturo Pomar, Lothar Schmid, and Peter Swinnerton-Dyer. Possibles include Peter Clarke (if he played T.H. George) and Gudbjartur Gudmundsson (if still alive).

ADDENDUM: Barden has pointed out on the English Chess Forum that Mortimer also played Ossip Bernstein at Ostend 1907. Bernstein lost to Frans Kuijpers at the 1961 IBM tournament. That apparently makes Kuijpers, now 69, the youngest living MN3. Bernstein also played Bent Larsen, who died earlier this month, twice at the 1954 Olympiad.

UPDATE: I have stricken through the names of MN3s who have died since this was first posted.

Saturday, August 14, 2010

The greatest swindle ever



You may have heard Evans-Reshevsky, U.S. Championship 1963-64 called "The Swindle of the Century". It's a nice one, to be sure, but Marshall-Marco, Monte Carlo 1904 is better. (Bizarrely, that game does not appear in Marshall's mistitled 1914 book Marshall's Chess "Swindles".) Keres-Fischer, Curacao 1962 is also an amazing save, but I wouldn't call it a "swindle" since Fischer (somehow) had no way to win after 72.Qe5!!

By the way, who christened Evans-Reshevsky "The Swindle of the Century"? Why, that would be Larry Evans himself, annotating the game in American Chess Quarterly, Vol. 3, No. 3 (Winter 1964), of which he was the editor-in-chief.

But the real Swindle of the (20th) Century, indeed the greatest swindle ever to date, is the amazing one that Larry Christiansen pulled off against Alexander Beliavsky at Reggio Emilia 1987-88. From a busted position, Christiansen sacrificed a knight to expose Beliavsky's king, then offered four pseudo-queen sacrifices in an attempt to get a perpetual check. Beliavsky thwarted Christiansen's attempts and repeatedly hammered Christiansen's king. Finally, on the brink of victory, Beliavsky fell for Christiansen's last trick.

After 29.Bc4, Christiansen had a lost position against the No. 5 player in the world. His f-pawn is under attack, but passive defense is hopeless, e.g. 29...Nh6 30.Qb6 winning the c-pawn (30...Qd7 31.Nxf7!). So Christiansen ignored White's threats and dove in with 29...Nxf2!? 30.Kxf2 Ra3! After 31.Bxf7+ Kg7 32.Qe6, he went after White's king with 32...Ra2+. Robert Byrne observed in The New York Times that after 33.Qxa2 Rxa2+ 34.Bxa2 Ng4+ 35.Kg1 Qa7 36.Bb1 Qa3 37.Bd3 Qb2 38.Rc2 Qd4+, "White will experience difficult technical problems." Instead, the game continued 33.Kg1 R8a3!, hoping for 34.Qxe7? Rxg3+ and the rook gives perpetual check along the third rank. Nor was 34.Kh1 Rxg3! 35.Qxa2 Ng4! appealing for White. Beliavsky preferred 34.Ne8+! Now 34...Nxe8? 35.Qxg6+ mates next move, and there is no perpetual check after 34...Qxe8? 35.Bxe8 Rxg3+ 36.Kh1. Undeterred, Christiansen played 34...Kh6! 35.Nxf6 35.Qxe7 Rxg3+ or 35.Qxf6 Qxf6 still leads to perpetual check. 35...Rxg3+ 36.Kh1 Qxf7! Offering the queen a third time, again hoping for perpetual check after 37.Qxf7? Rh3+ or 37.Ng8+? Qxg8! 37.Rd7! White offers his own queen sacrifice: if 36...Qxe6, 37.Rh7#! Another clear win was 37.Ng4+! hxg4 (37...Kg7 38.Qxe5+ is even worse) 38.Qxf7 Rh3+ 39.Kg1 Rg3+ 40.Kf1! Rf3+ 41.Qxf3, leaving White a rook up. 37...Qxf6! Black's last gasp, offering the queen yet a fourth time. 38.Qxf6?? White thinks that he can finally take the queen safely, since now there is no perpetual. White wins after 38.Rh7+! Kxh7 39.Qxf6 Rh3+ 40.Kg1 Rg3+ 41.Kf1 Rh3 41.Qe7+ Kh6 (41...Kg8? 42.Qe8+ Kh7 43.Qd7+ wins the rook) 42.Qg5+ Kh7 43.Kg1 Raa3 44.Kg2. He was rudely awakened by Christiansen's 38...Rh2+! A finish reminiscent of Evans-Reshevsky: Christiansen sacs his remaining heavy pieces and goes for stalemate. The players agreed to a draw in light of 39.Kxh2 Rg2+! 40.Kh3 Rg3+! 41.Kh2 Rg2+! 42.Kh1 Rg1+!, when Black draws by perpetual check or stalemate.

For another amazing Christiansen swindle, see Burden-Christiansen, Las Vegas 1992, wherein Christiansen spots a master a queen for a knight, and wins!

Tuesday, July 27, 2010

A busted line

The move 3...f5?! in Owen's Defense (1.e4 b6 2.d4 Bb7 3.Bd3 f5?!) has been played and analyzed for almost 400 years. Greco-NN, Rome 1619 quickly concluded 4.exf5! Bxg2 (the point, trapping White's rook - if Black can survive White's attack) 5.Qh5+ g6 6.fxg6 Nf6?? 7.gxh7+! Nxh5 8.Bg6#. Staunton pointed out the improvement 6...Bg7! in The Chess-Player's Handbook (1847), page 379. He analyzed 7.gxh7+ Kf8 8.hxg8(Q)+ Kxg8 9.Qg4 Bxh1 10.h4 e6 11.h5, with advantage to White. Over 120 years later, 10...Bd5!? (Schmid-Vitolinsh, Latvia 1969) and 10...Qf8! 11.h5 Qf6 12.h6 Rxh6 (Hendler-Radchenko, Kiev 1970) were shown to be playable for Black. Similarly, in Lombardy-Regan, U.S. Open 1974, the 15-year-old Regan got the advantage after 10.Nc3 Qf8! 11.Be3 Qf6 12.h3 Qh4 12.Qg6 Nc6. (Lombardy demonstrated why he's a grandmaster and we're not, holding a draw in an ending two(!) exchanges down.)

Alas for Black, later analysts convincingly refuted 3...f5? in two different and surprising ways. Accordingly, the French GM Christian Bauer now calls the move "simply suicidal" (Play 1...b6, p. 5). In both lines, White disdains the immediate capture on g8 and instead attacks Black's king. It is important to learn at least one of these busts so that you can refute 3...f5? if someone plays it against you.

In the mid-1970s, F.A. Spinhoven of the Netherlands found 8.Nf3!! Nf6 (8...Bxf3? 9.Qxf3+ and 10.Qxa8; 8...Bxh1 9.Ne5! Bxe5 10.dxe5 Bd5 11.hxg8(Q)+ Kxg8 12.Qg6+ Kf8 13.Bh6+) 9.Qg6! and now (a) 9...Bxh1 10.Bh6! Rxh7 11.Ng5! Bxh6 12.Nxh7+ Nxh7 13.Qxh6+ with a crushing attack or (b) 9...Bxf3 10.Rg1! Rxh7 11.Qg3!! ("the soul" of 8.Nf3!! - Soltis) Be4 12.Bxe4 Nxe4 13.Qf3+ and now White is a pawn up, with a winning attack, after 13...Kg8 14.Qxe4, or an exchange up with an easily winning position after 13...Nf6 14.Qxa8 Rxh2 15.Bf4 Rh4 16.Qg2 Rg4 17.Qh2 (Boris Avrukh, 1.d4, Volume Two, p. 551).

In 1982, Guido den Broeder found an alternative refutation, the stunning 7.Qf5!! Nf6 8.Bh6!! Bxh6 9.gxh7 Bxh1 10.Qg6+ Kf8 11.Qxh6+ Kf7 12.Nh3! and Black was quickly annihilated in den Broeder-Wegener, corr. 1982. See Wikipedia for more details. John Watson writes that den Broeder's 7.Qf5!! and Spinhoven's 8.Nf3!! both win, but that the analysis of 7.Qf5 "is complicated," while 8.Nf3 "is clearer" (Mastering the Chess Openings, Volume 4, p. 96).

Here is an online game I played yesterday. Unfortunately for my opponent, I was able to remember Spinhoven's analysis. Black played the latter part of the game weakly, but had an objectively lost position in any event.

Sunday, December 6, 2009

Chess in Wikipedia

A chess-related article is on the Main Page of Wikipedia as Today's Featured Article for only the fourth time ever. The article, about the obscure 19th-century chess player and writer George H.D. Gossip, was almost entirely written by me. After today, the article will be available only at its usual location in the online encyclopedia.